AC-DIMM

Author:

Guo Qing1,Guo Xiaochen1,Patel Ravi1,Ipek Engin1,Friedman Eby G.1

Affiliation:

1. University of Rochester, Rochester, NY

Abstract

With technology scaling, on-chip power dissipation and off-chip memory bandwidth have become significant performance bottlenecks in virtually all computer systems, from mobile devices to supercomputers. An effective way of improving performance in the face of bandwidth and power limitations is to rely on associative memory systems. Recent work on a PCM-based, associative TCAM accelerator shows that associative search capability can reduce both off-chip bandwidth demand and overall system energy. Unfortunately, previously proposed resistive TCAM accelerators have limited flexibility: only a restricted (albeit important) class of applications can benefit from a TCAM accelerator, and the implementation is confined to resistive memory technologies with a high dynamic range ( R High /R Low ), such as PCM. This work proposes AC-DIMM, a flexible, high-performance associative compute engine built on a DDR3-compatible memory module. AC-DIMM addresses the limited flexibility of previous resistive TCAM accelerators by combining two powerful capabilities---associative search and processing in memory. Generality is improved by augmenting a TCAM system with a set of integrated, user programmable microcontrollers that operate directly on search results, and by architecting the system such that key-value pairs can be co-located in the same TCAM row. A new, bit-serial TCAM array is proposed, which enables the system to be implemented using STT-MRAM. AC-DIMM achieves a 4.2X speedup and a 6.5X energy reduction over a conventional RAM-based system on a set of 13 evaluated applications.

Funder

International Business Machines Corporation

Division of Computing and Communication Foundations

New York State Office of Science and Technology

Qualcomm

Cisco Systems

Samsung

Publisher

Association for Computing Machinery (ACM)

Reference47 articles.

1. Design Compiler Command-Line Interface Guide. http://www.synopsys.com/. Design Compiler Command-Line Interface Guide . http://www.synopsys.com/.

2. Free PDK 45nm open-access based PDK for the 45nm technology node. http://www.eda.ncsu.edu/wiki/FreePDK. Free PDK 45nm open-access based PDK for the 45nm technology node. http://www.eda.ncsu.edu/wiki/FreePDK.

3. Advanced Micro Devices Inc. AMD64 Architecture Programmer's Manual Volume 2: System Programming 2010. Advanced Micro Devices Inc. AMD64 Architecture Programmer's Manual Volume 2: System Programming 2010.

4. Hybrid CMOS/nanodevice circuits for high throughput pattern matching applications

Cited by 41 articles. 订阅此论文施引文献 订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献

1. Computing with nonvolatile memories for artificial intelligence;Semiconductor Memories and Systems;2022

2. OrderLight: Lightweight Memory-Ordering Primitive for Efficient Fine-Grained PIM Computations;MICRO-54: 54th Annual IEEE/ACM International Symposium on Microarchitecture;2021-10-17

3. Compact Model of Magnetic Tunnel Junctions for SPICE Simulation Based on Switching Probability;IEICE Transactions on Electronics;2021-03-01

4. Analog content-addressable memories with memristors;Nature Communications;2020-04-02

5. Energy-Efficient Stochastic Computing with Superparamagnetic Tunnel Junctions;Physical Review Applied;2020-03-05

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3